Cycling race predictor · one-day race

Tre Valli Varesine settled

One day, one result: the production stack priced the whole field under 1 course reading, against the market's line where one was captured. A one-day race is its own best reference — the main reading is built from this race's own recent editions.

Raced 6 October 2026 1 reading 170 riders priced models retrained 72bc1bd857 market snapshot no capture

The fork

1 reading of the same road

Course profile is an input the model cannot see from a map — it is set from reference races. The main reading averages this race's own last editions, so the terrain and the kind of rider it selects come from the race's own history; any alternate reading swaps in an archetype set to ask what a different day would do to the same field.

Own history

Tre Valli Varesine on its own recent editions
sprint0.01
mountain0.12
hilly0.04
rolling0.72
break0.11
TT0.00
References: Tre Valli Varesine (2025); Tre Valli Varesine (2023); Tre Valli Varesine (2022); Tre Valli Varesine (2021); Tre Valli Varesine (2019)
The race read on its own recent history: the profile is the average of this race's last 5 editions, so the terrain and the kind of rider it selects come from the race itself rather than from an archetype.
Top 5 (blend): Pidcock Tom 6.5%; Grégoire Romain 5.3%; Tiberi Antonio 3.4%; Martinez Lenny 3.4%; Scaroni Cristian 3.2%

The board

Win probabilities across the heads

Blend is the production price (cascade × LambdaRank, DNF-adjusted at the pricing layer); Cascade and LambdaRank are the two underlying heads shown for the main reading. The market columns are the devigged win line, present as the information benchmark the model is scored against.

Read "Max edge" as directional, not as a price. The blend is a set of per-rider marginals and does not sum to one book — across this field each reading totals own history 79% — while the devigged market column sums to exactly 100%. Putting the two side by side flatters the model, and Max edge is inflated by about that factor. Of our heads only the LambdaRank one is natively book-preserving.

RiderTeamBlendCascadeLambdaRankMarket oddsMarket devigMax edge
Pidcock TomPinarello-Q36.5 Pro Cycling Team6.5%2.8%9.2%n/a––
Grégoire RomainGroupama-FDJ United5.3%6.1%6.6%n/a––
Tiberi AntonioBahrain Victorious3.4%2.3%4.6%n/a––
Martinez LennyBahrain Victorious3.4%2.1%4.7%n/a––
Scaroni CristianXDS Astana Team3.2%2.3%4.3%n/a––
McNulty BrandonUAE Team Emirates-XRG2.6%4.7%2.8%n/a––
Pellizzari GiulioRed Bull-BORA-hansgrohe2.3%3.1%2.7%n/a––
Johannessen Tobias HallandUno-X Mobility2.2%2.4%2.7%n/a––
Hirschi MarcTudor Pro Cycling Team2.2%2.1%2.8%n/a––
Lapeira PaulDecathlon CMA CGM Team1.9%3.2%2.1%n/a––
Cosnefroy BenoîtUAE Team Emirates-XRG1.8%2.9%2.0%n/a––
Rondel MathysTudor Pro Cycling Team1.5%1.0%2.1%n/a––
Wellens TimUAE Team Emirates-XRG1.5%2.1%1.8%n/a––
Piganzoli DavideTeam Visma | Lease a Bike1.5%1.7%1.8%n/a––
Vendrame AndreaTeam Jayco-AlUla1.5%2.2%1.7%n/a––
Van Gils MaximRed Bull-BORA-hansgrohe1.3%2.4%1.4%n/a––
Mas EnricMovistar Team1.3%1.2%1.6%n/a––
Buitrago SantiagoBahrain Victorious1.2%0.8%1.6%n/a––
Jorgenson MatteoTeam Visma | Lease a Bike1.2%2.9%1.0%n/a––
Crescioli LudovicoTeam Polti VisitMalta1.1%0.4%1.6%n/a––
August AJNetcompany INEOS1.1%0.3%1.6%n/a––
Honore MikkelEF Education-EasyPost1.0%1.0%1.3%n/a––
Tre Valli Varesine on its own recent editions — top 30, all heads
#RiderBlendCascade LambdaRankPodiumTop 15 Market odds
1Pidcock Tom6.5%2.8%9.2%12.2%45.0%n/a
2Grégoire Romain5.3%6.1%6.6%18.4%44.4%n/a
3Tiberi Antonio3.4%2.3%4.6%8.4%26.3%n/a
4Martinez Lenny3.4%2.1%4.7%7.9%25.2%n/a
5Scaroni Cristian3.2%2.3%4.3%9.5%41.7%n/a
6McNulty Brandon2.6%4.7%2.8%11.0%30.3%n/a
7Pellizzari Giulio2.3%3.1%2.7%8.7%35.7%n/a
8Johannessen Tobias Halland2.2%2.4%2.7%9.0%33.8%n/a
9Hirschi Marc2.2%2.1%2.8%7.5%36.8%n/a
10Lapeira Paul1.9%3.2%2.1%10.7%37.0%n/a
11Cosnefroy Benoît1.8%2.9%2.0%11.1%37.8%n/a
12Rondel Mathys1.5%1.0%2.1%5.1%26.0%n/a
13Wellens Tim1.5%2.1%1.8%8.8%34.3%n/a
14Piganzoli Davide1.5%1.7%1.8%7.4%34.9%n/a
15Vendrame Andrea1.5%2.2%1.7%9.7%44.4%n/a
16Van Gils Maxim1.3%2.4%1.4%9.8%35.0%n/a
17Mas Enric1.3%1.2%1.6%4.9%16.3%n/a
18Buitrago Santiago1.2%0.8%1.6%2.5%16.3%n/a
19Jorgenson Matteo1.2%2.9%1.0%7.0%31.8%n/a
20Crescioli Ludovico1.1%0.4%1.6%2.2%28.6%n/a
21August AJ1.1%0.3%1.6%0.7%6.4%n/a
22Honore Mikkel1.0%1.0%1.3%4.2%18.4%n/a
23Tulett Ben1.0%2.0%1.0%8.2%27.9%n/a
24Arensman Thymen1.0%1.5%1.1%2.8%11.0%n/a
25Sheffield Magnus0.9%1.5%1.1%5.1%17.5%n/a
26Simmons Quinn0.9%1.4%1.0%5.8%25.1%n/a
27Ulissi Diego0.9%0.8%1.2%4.3%20.9%n/a
28Narvaez Jhonnatan0.9%1.0%1.1%2.4%21.9%n/a
29Pinarello Alessandro0.8%0.4%1.1%1.9%20.4%n/a
30Omrzel Jakob0.8%0.8%1.0%2.2%15.0%n/a

After the race

Result — won by Widar Jarno

PosRiderTeamOwn history win p
1Widar JarnoLotto-Intermarché0.6%
2Pidcock TomPinarello-Q36.5 Pro Cycling Team6.5%
3Jorgenson MatteoTeam Visma | Lease a Bike1.2%
4Simmons QuinnLidl-Trek0.9%
5Hirschi MarcTudor Pro Cycling Team2.2%
6Bilbao PelloBahrain Victorious0.2%
7Scaroni CristianXDS Astana Team3.2%
8Crescioli LudovicoTeam Polti VisitMalta1.1%
9Narvaez JhonnatanUAE Team Emirates-XRG0.9%
10Gualdi SimoneLotto-Intermarché0.2%

Scoring — probability given to the actual winner

ReadingBlendCascade LambdaRankMarket devig
Own history0.62%0.32%0.86%–